Virtual Depository: Arizona Project

1. Virtual Depository: Arizona Project Final Report and Recommendations October 2003 Atifa Rawan University of Arizona Library

2. Project Purpose • Transform the University of Arizona’s Library’s existing depository collection into an all-electronic depository. • Assess the implications of selecting government information resources within the depository profile only in electronic form, which had not been previously permitted.

3. Project Partners • University of Arizona Library • University of Arizona School Of Information Resources and Library Science (SIRLS) • Library Programs Service of the United States Government Printing Office (GPO)

4. Why This Project? • The idea arose out of discussions at the fall 2001 meeting of the Depository Library Council to the Public Printer. • The rapid growth of federal government information becoming available on the Internet has consequences for Federal Depository Library Program (FDLP). • Since 1993, GPO and its Depository Council has been preparing reports and assumptions concerning GPO’s transition to a more electronic environment.

5. Why was the pilot developed?GPO’s Perspective • The proliferation of government resources on the internet is forcing depositories to become more integrated with the larger library community. • The pilot project was intended to assess implications for users of government documents and the management of federal depository information in an electronic environment

6. Why this Project?University of Arizona’s Perspective The Pilot supported our Library’s strategic goals Access goals as a driver • We provide integrated and convenient access to information from multiple perspectives and sources in formats that match customer needs. • A guiding principle is that users should be able to access information self-sufficiently from any place at any time. Space as driver • Shelving space is a major concern at the University of Arizona Library. Reduced budgets will make interim collection space options through compact storage difficult to achieve.

7. Pilot Project Activities Activities included: Create Project Team Selection & Modification Identify Electronic (EL) Docs Needs Assessment Outcomes/ Learnings Recommend/ Commend

8. Key Activities of the Project:Revise the Library’s Selection Profile UA’s Depository Profile 2002-03 Item #’s selected 3586 59% Items #’s not selected 2472 41% Total possible 6058 100%

9. Revise the Library’s Selection Profile Exceptions to the Electronic-only Profile 1. Identified a short list of titles to select in both electronic and tangible formats 2002/03 - 25 exception titles 2003/04 - 23 exception titles Exception Titles 03/04 – 23 2. The Library elected to receive selected maps and datasets in their original tangible format

10. Creation of two profiles1. UNIV of ARIZ/VLC PILOT PROJ - - Library 1070. This was an inactive profile based on the Library’s 01/02 selection profile. this profile included 3,292 (54.41%) items out of possible selections of 5,941 of the total items offered

11. 2. University of Arizona -- Library 0023 . This was an active profile based on 02/03 item selection profile which included many electronic titles.

12. Identification of Electronic Federal Government Documents • During the pilot, the UA Library submitted 3,288 titles and non-working URLS to lostdocs@gpo.gov. Subsequently, these were reviewed for inclusion in the program by LPS staff in association with their Electronic Working Group project. • Total number of items checked to verify full-text EL documents from variety of government sites by UA Library staff were 3,509 items (6,423 titles). We used our 2002/03 profile as our basis.

13. Identification of Electronic Federal Government Documents • 25% of these titles were searched more than once. On the average, it took the Library’s experienced intern 15 minutes per title to search and to verify the exact title or series. • Total number of files submitted to GPO lostdocs were 26 files. The largest file contained 560 titles. The GPO has discontinued 12 titles. Currently, the Library and the GPO are collaborating in the continuation of this project. The Library will to continue to verify titles for lostdocs. As part of this collaboration, the GPO has hired two Library Science graduate assistants (Knowledge River) to help the Library with this task. See Appendixes A-C for detailed metrologies

14. Needs Assessment and Data Gathering • The UA Library partnered with SIRLS to develop and implement assessment tools. The development of an instrument to assess the transition process and use of the depository resources in the online-only format began in September 2002. • The partnership completed pre and post assessment of students in two Journalism undergraduate classes and a large Political Science undergraduate class • An assessment tool for IC service providers was designed in fall 2002. Data were gathered for two weeks and were analyzed.

15. Pilot Assessment Continued… • Completed two information sessions given about the project for all reference providers • Completed a focus group with one faculty in Journalism and one from Political Science department • Awareness session for Tucson Pima Public Library and the Arizona Health Sciences staff • Worked with a local company “Informatouch” and conducted a survey of users satisfaction with electronic government resources in April and May, 2003 at Main and Science libraries

16. Assessment Purpose • Fall semester: conducted initial surveys of staff and students that would help us design a survey instrument for use in the spring • Spring semester: discovered barriers to and advantages of Web-based government information from perspectives of library staff and customers • See Survey Results

17. Learnings/Outcomes • In general, undergraduate students were savvy about surfing the net. They were very comfortable using search engines. However, they did not understand search concepts. Rather than relying on one single method of finding government information, students tended to try different approaches. • Library staff were usually looking for a specific source and were more aware of search concepts. • Students, for the most part, were quicker in making a decision. The staffs were more concerned with finding the right answer, and it took them longer

18. More Learnings/Outcomes • Because of the way PURL searches go to agency sites, it was not feasible to count specific item numbers accessed by customers. This impaired our ability to keep statistics. • Students indicated the biggest advantage of using web-based government information was their ability to access information anytime without going to the Library. They also liked finding lots of information on a topic they were researching. • Almost all users indicated that they did not encounter difficulty when using web-based government information.

19. More learnings… • All survey users indicated that they prefer using the online sources and they were satisfied with e-documents • It is important to provide instruction/training for students, faculty, and staff • Existence and maintenance of online catalog records with working PURLs is essential. Additional, locally created tools are important supplements to online catalog

20. Recommendations/Issues • We fully support the implementation of a virtual depository for our Library starting 9/1/03. Our decision is based on input from our customers, their pattern of use and level of satisfaction, our Library’s collection development policy, and the GPO’s visions and directions. • In order to move ahead with this implementation, we worked with the Library Information Resources Council and received their endorsement in changing our collection development policy to reflect this shift, and revised and updated our depository profile for 2003-2004 selection cycle to reflect the Library’s mission and vision.

21. Recommendations -Cont. We are committed to continue: • our systematic and consistent procedures for identifying fugitive electronic publications and bringing them into the FDLP • to provide fuller access for those who rely on the Library’s online catalog to find and to locate government information • our partnership and collaboration with GPO and the State Library • to promote instruction and information literacy for government publications and market our virtual depository collection

22. Project Assessment and Outcomes This pilot served an important purpose for the University of Arizona Library: • It helped the pilot project team assess the feasibility of beginning the transition to a virtual depository collection. • The Library, after a full year of assessment, decided to become a virtual depository for those materials available and archived by GPO. • Under its newly revised collection development policy, the Library has begun broadly substituting the EL format in place of the tangible, particularly when doing so aligns with the Library's own understanding of its unique position and philosophy.

23. Project Assessment and Outcomes • The pilot study attempted to take users’ characteristics, usage patterns, community needs, research requirements, and collection development policy factors into account This was done by directly questioning library staff and users and by relying on the knowledge of UA Library information resource managers and of other librarians outside the UA Library system. • Some of the responses gathered during the pilot year suggest that research is needed to understand why customer satisfaction with Web-based government information is not 100% and what a single selective depository might do to increase satisfaction levels.

24. Project Assessment and Outcomes… • Other comments from customers suggest that more vigorous promotion and instruction might be helpful for the various constituencies using government information. Also needed is further investigation of faculty perceptions of Web-based government information, perhaps through group interviews rather than surveys.

25. Project Assessment and Outcomes… • During the pilot year, the Library closely monitored customers’ needs and responses. • The first step was to revise the Library’s federal depository collection development policy. • The next step was to receive support and endorsement from the Library Information Resources Council, the Library information resource managers, and the Library Technical Services and Archiving Processing Team.

26. Project Assessment and Outcomes… • The Pilot provided time and methodology for the Library staff to make changes based on data gathering and users’ needs assessment.. • Among the information desk staff and students who answered questions, there is a comfort level with, and in many cases a preference for, government information in electronic form. The trend was already clear, but the study helped document it for the UA Library.

27. Project Assessment and Outcomes… • Throughout the pilot year, the library had two back-up systems in place in case the project created unforeseen problems with electronic access. • Clearly the transition to virtual depository status requires continued monitoring, staff commitment and assessment to ensure that the library's staff and customers are benefited rather than disserved by the decision. But equally clear, in the case of the UA Library, is the feasibility of moving forward with the transition.

29. Savings The Library’s space savings include: • approx. 190 linear feet of shelving space or 63 shelves • one microfiche cabinet drawer • at least one map cabinet

30. Staff Time We saved staff time by not processing and shelving 400 boxes of paper, 200 separates (individual paper titles), 100 packages of microfiche, and 100 cylinders of maps. This freed up staff and students to perform other tasks in the Library. The items processing time has been cut to 10% by technical processing staff.

31. Conclusion Entering into the UA Library's general policy of preferring electronic over tangible formats is the issue of the continuing availability of the information resources. In cases where the publishers and distributors of electronic information resources take responsibility for archiving those resources, the library prefers to provide access to the current and archived information rather than itself creating and storing archival back-ups. In GPO's case, the decade-long transition to a more electronic system created the possibility of the virtual depository being implemented at UA. The UA Library interpreted the availability of government information on the Internet not as a reason to leave but as an opportunity to explore a new model.

32. Questions/Discussions

33. Sources Searched___Appendix A • The minimum information required for sending to lostdocs included the full title or series of the work, item number, full URL associated with a series or titles (e.g http://pubs.usgs.gov/pp/p1657/), any associated PURL found, and possible OCLC numbers. In general, records with only one issue of a serial title or general publications, and series if they were problematic and too broad were not included. General publications and handbooks, manuals, guides were not searched. Full content was included - not just partial content on the agency sites. • Sites searched included: • GPO Access - http://www.access.gpo.gov/su_docs/index.html; WorldCat – Database available through Sabio - Looking for cataloging location such DLC or GPO and looking for PURL addresses; Google - advanced search - http://www.google.com; Federal web sites hosted on GPO access - http://www.access.gpo.gov/su_docs/sites.html; catalog of US Government Publication - http://www.access.gpo.gov/su_docs/locators/cgp/index.html, various universities’ government agencies ;SearchGov - http://www.searchgov.com/; FirstGov - http://www.firstgov.gov/

34. Creation of Master Working File and a Database_Appendix B • Created an internal master working file using excel. The file included all Library profile items. The file was color coded to show the status of each record. This file included full information regarding the sudoc, title, item number, format, frequency, PURL/URL, and comments. The file was updated whenever the following changes occurred: • When documents were sent to GPO, Lostdocs@gpo.gov • The List of Classes – All EL, including multi-format publications were received from GPO • List of discontinued items were found in the Administrative Notes Technical Support • Dead PURL links were found • The Library dropped any items from its profile • The most current file was uploaded into the Library’s Government Documents web page (http://www.library.arizona.edu/library/teams/sst/pol/guide/gpo-profile.htm) • The Library created a separate database that included item numbers, SUDOC numbers and titles. The electronic titles were hyperlinked. These lists were uploaded onto the Government Documents web page at: http://www.library.arizona.edu/library/teams/sst/pol/guide/govdocs.html • The list was maintained and updated regularly. The final item list was given to the Library’s Technical Services and Archival Processing Team (TSAP) to be forwarded to Marcive to upload depository records to the Library online catalog. The TSAP staff are linking directly to the online resources on the agency sites or archived copies on the GPO servers.

35. General Observation and Tips in Searching-Appendix C • Seeing more of the Interior publications in the Catalog of Government Publications on the GPO web site • The Bureau of Indian Affairs (BIA) site was down during the pilot due to Cobell vs. Norton, litigation over the Indian trust fund. • Searching could slow when using sites on Internet Explorer. Old Netscape browsers are more problematic retrieving full-text documents • In general, it is easy to navigate in most agency sites. However, there were few redesigned web sites that were not user friendly and finding full-text publications was challenging. Perseverance and persistence were needed to continually find publications. Difficult sites for navigation included Department of Interior, Agriculture Agencies such as the National Park Service, NOAA, Military sites where some documents were removed, and Department of Education

36. Continued -Appendix C • The most common error was when there was a purl accessed through GPO that was to be for a series, but it was only one of the documents in the series. There was no way to go to the series from the single purl and further research was needed. • There were instances, for example, of an annual report that was updated every year but the purl was for 1998 • Some agencies also made it easier to find documents because they have links on their homepages to such things as “Document Libraries”, which were set up in easy to search formats with documents available only 2 clicks into the website. Also helpful in some agencies were the “Site Map” links which broke down busy and/or “deep” websites into a much more manageable form to search for documents. • Some agencies have also developed specific databases to search for documents and reports. For example the Department of Transportation has developed a researchable database called TRIS (http://199.79.179.82/sundev/search.cfm) in order to find transportation research documents